vina-diesel - an automated autodock-vina wrapper

Autodock vina is a molecular docking command line program. vina diesel is a simple python wrapper that automates some of it.

original paper

  title={AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading},
  author={Trott, Oleg and Olson, Arthur J},
  journal={Journal of computational chemistry},
  volume={31},
  number={2},
  pages={455--461},
  year={2010},
  publisher={Wiley Online Library}
}

how vina-diesel works

vina works on pdbqt files, which vina-diesel creates in a tempfile using openbabel. vina-diesel requires a pdb file (receptor) and a smiles string (ligand). Unless otherwise specified, all non-protein atoms are removed from the receptor structure when a vdsl.protein object is initialized (The first chain is also selected by default).

vina requires a box to be drawn around the area of the receptor to be docked. In vina-diesel this is specified using the residue numbers of the target site which can be found by playing with a 3D structure viewer like pymol. If the residue numbering in a structure is different to how you like (e.g. you're using a truncated or extended sequence) then provide the full sequence to vdsl.protein to align that sequence to the structure and you'll be able to use your sequence numbering.

vina-diesel calls the vina binary as a subprocess and stores the results in a results object, which can be used to save the poses and energies or access a pandas DataFrame of the pose energies and a biopandas.pdb object containing the coordinates. The results object is a little bit scrappy, sorry about that.

simple example

import vdsl

p = vdsl.protein('3b4y.pdb') ###
results = p.dock('[O-][N+](=O)c1cn2C[C@@H](COc2n1)OCc3ccc(OC(F)(F)F)cc3', 
    target_sites = [38, 44, 175, 176, 199, 252, 256, 261, 283, 311]) # 

results.save('thatwaseasy')

another example

import vdsl

MY_SEQUENCE = 'MTAJSKNFHASBOYBRWBRHA...' # keyboard mash
COFACTORS = ['HEM', 'FAD'] # cofactors to keep -- names as they appear in the pdb file

p = vdsl.protein('3b4y.pdb',
		seq = MY_SEQUENCE,
		keep = COFACTORS)
results = p.dock('[O-][N+](=O)c1cn2C[C@@H](COc2n1)OCc3ccc(OC(F)(F)F)cc3', 
	    target_sites = [38, 44, 175, 176, 199, 252, 256, 261, 283, 311], # 
		exhaustiveness=16)  # integer 1-16 (coarse-fine grained)

results.save('thatwaseasy')

pritn(results.scores) # pd.DataFrame

installation

warning I've only tested this using conda - your best bet is to clone the environment in env.yml and pip install. I had issues with the different versions of openbabel in the past so it's potentially risky business to roll your own installation.

another warning I've also only tested this on linux distros (ubuntu, debian, arch) - let me know if something weird happens

• clone the repository: git clone https://github.com/UoMMIB/vina-diesel.git
• move in cd vina-diesel
• clone the environment with conda conda env create -f env.yml
• ACTIVATE conda activate vdsl
• install pip install . (from the root with setup.py)

done. test with test/test.py if you like